pekat-vision-sdk 1.4.4

A Python module for communication with PEKAT VISION

PEKAT VISION SDK

A Python module for communication with PEKAT VISION

Requirements

• numpy
• python-opencv (optional)
• PEKAT VISION 3.10.2 or higher

Installation

Type pip install pekat-vision-sdk

Usage

import cv2
from PekatVisionSDK import Instance

p = Instance('C:\\Users\\Peter\\PekatVisionProjects\\test_project')

# response - only context
context = p.analyze('path_to_image.png')

# response - heatmap and context
img_heatmap, context = p.analyze('path_to_image.png', response_type='heatmap')

# response - annotated image and context
img_annotated, context = p.analyze('path_to_image.png', response_type='annotated_image')

p.stop()

You cand send image in numpy array:

import cv2
img_np = cv2.imread('path_to_image.png')

p.analyze(img_np, response_type='annotated_image')

Access to already running PEKAT VISION instance

p = Instance(port=8100, host='192.168.10.0', already_running=True)

If secure image analyze is enabled for the Pekat project, you need to use the 'api_key' parameter with the corresponding API key that was generated for the project:

p = Instance(port=8000, host='127.0.0.1', already_running=True, api_key='write_your_api_key_here')

Remote analyzer enables you to connect to a remotely running PEKAT VISION. It is possible to connect from multiple PCs simultaneously. PEKAT VISION behaves as a server automatically.

Multiple cameras

# start projects
p1 = Instance('C:\\Users\\Peter\\PekatVisionProjects\\project_camera_1')
p2 = Instance('C:\\Users\\Peter\\PekatVisionProjects\\project_camera_2')
p3 = Instance('C:\\Users\\Peter\\PekatVisionProjects\\project_camera_3')


# response - only context 
# in loop
context1 = p1.analyze('path_to_image_camera_1.png')
context2 = p2.analyze('path_to_image_camera_2.png')
context3 = p3.analyze('path_to_image_camera_3.png')